Adversarial Examples Detection With Bayesian Neural Network

Yao Li

doi:10.1109/TETCI.2024.3372383

Adversarial Examples Detection With Bayesian Neural Network

2024-10-01

1

OA

AI

DOI：10.1109/TETCI.2024.3372383

Original

Original request for help

Save

Abstract

En 中文

In this paper, we propose a new framework to detect adversarial examples motivated by the observations that random components can improve the smoothness of predictors and make it easier to simulate the output distribution of a deep neural network. With these observations, we propose a novel Bayesian adversarial example detector, short for BATer, to improve the performance of adversarial example detection. Specifically, we study the distributional difference of hidden layer output between natural and adversarial examples, and propose to use the randomness of the Bayesian neural network to simulate hidden layer output distribution and leverage the distribution dispersion to detect adversarial examples. The advantage of a Bayesian neural network is that the output is stochastic while a deep neural network without random components does not have such characteristics. Empirical results on several benchmark datasets against popular attacks show that the proposed BATer outperforms the state-of-the-art detectors in adversarial example detection.

Keywords:

Bayes methods

Artificial neural networks

Training

Robustness

Gaussian distribution

Detectors

Computer architecture

Adversarial example

Bayesian neural network

deep neural network

detection

Adversarial Examples Detection With Bayesian Neural Network

Abstract

Journal

Organization